AI-Enabled Intelligent Intrusion Detection Framework Using Artificial Neural Networks for Secure and Sustainable Networked Systems

Abstract

The explosion of cloud computing, online services, and interlinked digital services has contributed to increased susceptibility of modern networks to cyber-attacks. Traditional Intrusion Detection Systems (IDS) detect attacks by utilising signature-based detection methods, which often fail to recognise novel or previously unrecorded attack patterns. To counter these shortcomings, the research will describe a sophisticated Artificial Neural Network (ANN) application, designed to not only improve the effectiveness of cyber security systems, but also boost the overall rate of threat detection. Proposed detection systems will improve cyber security by employing the ability of neural networks to learn patterns, and will therefore be able to evaluate and categorise network activity as being acceptable, or as representing a threat. The complete system will consist of a number of steps including, but not limited to, the acquisition of datasets, and the application of preprocessing.feature encoding, feature normalization and selection to improve data quality and minimize redundancy. It is a multilayer feedforward neural network model that is trained and tested over benchmark intrusion detection datasets against a number of attack types that include Denial-of-Service (DoS), Probe, Remote-to-Local (R2L), and User-to-Root (U2R) attacks. As demonstrated through experimental analysis, the proposed ANN model will achieve high precision and recall in addition to low false positive rate at 97.6 percent. Additional comparative study can show that ANN-based methodology outperforms other traditional machine learning algorithms, such as Decision Trees, Support Vector Machines, and Random Forest classifiers. The results show that neural network-based solutions can be useful in detecting complex intrusion patterns and making real-time network security in modern computing and cloud-based systems, with Internet of Things (IoT) networks.